Occupational therapy, like other health professions, is under increased scrutiny to justify reimbursement for services. In order for the profession to stay vital, research that investigates the effectiveness of occupational therapy treatments is critical. Research into treatment effectiveness supports a model of evidence-based practice that can improve quality of care and help define the unique roles of health professions in new areas.

One new area of evidence-based practice for health professions is prevention and wellness. There is a growing body of research on the impact of positive lifestyle factors on health. Engagement in physical activity has been shown to have physical and mental health benefits and to decrease the severity of chronic health conditions such as arthritis. For people with arthritis, participation in physical activity has been shown to reduce the rate of functional decline and to be protective against the development of problems in performing activities of daily living.

Despite the benefits of engaging in physical activity, people with arthritis are more inactive than people without arthritis. People with osteoarthritis have disease-specific problems that may create additional barriers to adopting a more active lifestyle. For example, people with osteoarthritis tend to have pain and fatigue symptoms that could be exacerbated by starting a structured exercise program. Other people may fear that exercise will harm their joints. In addition, there are often barriers in people’s homes and communities that make engaging in physical activity difficult.

Each of these factors were considered when developing the occupational therapist-led program – Activity Strategy Training – for people with osteoarthritis. Activity Strategy Training (AST) is a term we developed to help explain what occupational therapists provide in this program; that is, the occupational therapists-taught strategies on how to perform desired and necessary activities with respect to disease-related factors.

In this intervention, we wanted to address symptom management and promote physical activity for people with osteoarthritis. We purposely designed the study to be performed at places where barriers could be addressed and potential solutions could be discovered, such as in senior housing facilities. We chose a small group format with two occupational therapist leaders to facilitate peer-related problem solving and to practice the concepts that were taught.

The Aim to Heal

The program was designed to educate patients about protecting their joints, proper body mechanics, activity pacing, and decreasing environmental barriers. The education component included the discussion of current research evidence, which is similar to what has been done in other prevention programs. There was also a practice portion or didactic exercise in each session. For example, a session about body mechanics involved the practice of techniques for walking up and down stairs and getting in and out of a car.

After some sessions, participants were asked to complete homework assignments. For a session on activity pacing, participants completed a daily log of their physical activity and symptoms to help them become aware of how the two concepts are related. Participants could then examine when symptoms worsened during their day and during which activities.


Discussion on using the strategy of activity pacing not only included how to take breaks and slow down during activities, but also how to add in an activity to stay at a “constant” pace. In addition to the education, group discussion, and practice of techniques, there were home visits where one occupational therapist examined in-home barriers to activity and recommended adaptive equipment if needed.

The intervention involved eight sessions, two times a week for 90 minutes, conducted over a four-week period. In addition to the AST, a structured progressive resistive exercise program using ankle weights was incorporated into every session. The intervention emphasized the importance of building in lifestyle physical activity – for example, extra walks within or outside the housing facility – and the structured exercise program was presented as one way to help make physical activity engagement a habit.

Therefore, we felt it was necessary to provide a structured program that could be done easily indoors, using the handouts and weights provided. The amount of weight used at the baseline session was tailored for each participant, and people progressed slowly with attention to any new symptoms resulting from beginning and progressing the exercise regime.

In between the AST and the exercise component, a short break was given and light snacks were provided. The goal of the break was to increase socialization between group members and potentially enhance the peer support to engage in physical activity within each housing facility. The therapist leaders underwent an initial orientation and training with the principal investigator. The group leaders were given a detailed program manual that outlined goals for each session, the order and timing of each activity, and some scripted information.

A flip chart was used to convey education content and handouts were provided to participants, which were continuously added to each participant’s program folder. If participants could not attend a session, the leaders ensured that handouts were received by the non-attendees. After the eight sessions, two booster sessions were held over a 6-month period to reinforce key concepts of the program and address any issues that arose.
 
Research Design Considerations

The complete description and results of the pilot research study of the effectiveness of activity strategy training has been reported in Arthritis & Rheumatism, the official journal of the American College of Rheumatology. In brief, we performed a pilot randomized controlled trial to examine feasibility issues with administering the intervention and to determine if there were treatment effects.

Designing research to test occupational therapy effectiveness is challenging. One issue that can be difficult is determining the appropriate control group. We felt that it was important for both groups to receive the same exercise component. The exercise component of programs in our experience is critical to get people to enroll in a program like this. Then we could examine a direct comparison of the AST versus our control group, which we designed as health education.

An inert control group (a group not expected to impact the outcomes being measured) would be ideal because a main source of bias in these types of therapeutic trials is treatment time. Specifically, the more time spent in the intervention with the intervention providers produces more positive results. So an inert control group could control for treatment time, yet not be expected to affect the results.

In this study, we felt that the control group could potentially affect the outcomes being measured, but health education was probably the most appropriate control group given the information that may be typically provided to people with osteoarthritis. Similar to the AST group, the health education group participated in eight 90-minute sessions that included the progressive resistive exercise training. Participants also met in small groups with two group leaders and they received handouts.

The main difference in the groups was the information provided. Instead of focusing on problem solving and the barriers to physical activity, the health education group received information on the osteoarthritis disease process, discussed the importance of physical activity (without linking it to their personal barriers), and other information, such as nutrition and effectively communicating with healthcare providers. The information presented by the group leaders was taken primarily from publicly available osteoarthritis information from the Arthritis Foundation.

Another issue taken into consideration in designing the research study was how to randomize given that we would be performing two interventions at different sites. We had to decide whether we should randomize ‘by site’ in which the site itself would determine whether participants would receive either AST or health education, or randomize participants into each intervention within each site.

We chose to randomize participants to receive one of the two interventions within each site because randomizing by site required many more sites and an attempt to match sites by characteristics, such as size and minority representation. Randomizing participants to receive one of the two interventions within each site reduced the site effects and didn’t require site matching.

A main limitation of having participants receive different interventions within each site was the potential for contamination. That is, participants could be talking to each other about their intervention program, reducing effects produced by unique aspects of one of the interventions. Given that we were conducting a pilot study, we evaluated this potential problem by holding exit interviews with participants to assess for contamination effects. It appeared that there was low contamination in this study. Of the 47 participants who responded to the exit interview, only 3 reported discussing the content of the program with someone who was in the other intervention group.  

A third issue was how to measure physical activity, one of the primary outcomes. In older adults, self-report physical activity measures may not be ideal given problems with recall bias and memory, while more objective physical activity measures could provide important information on physical activity patterns that could be obtained easily. 

In this study, we decided to use both subjective and objective physical activity measures. The CHAMPS, a recall instrument designed for older adults was used, as well as objective physical activity assessment using wrist-worn accelerometers. We were interested in determining if we’d see similar trends from the different types of physical activity measures and examine the feasibility of using the objective measurement in this study. 
 
A final issue was attempting to examine potential mechanisms underlying the effectiveness of this occupational therapy intervention. We felt that participants in the AST intervention would be gaining confidence in abilities to manage their symptoms through practice, peer support, and discussion, so we identified arthritis self-efficacy as a potential mechanism of treatment – this outcome was measured by the Arthritis Self-Efficacy Scale.   

Study Results

The study sites included three senior housing facilities and one senior center in Southeastern Michigan. A total of 84 older adults were screened and 64 percent (n = 54) were randomized into one of the interventions. Reasons for ineligibility and dropouts are outlined in the original research study (Murphy et al., 2008).

Of participants, 89 percent had knee osteoarthritis with or without hip osteoarthritis. Participants had a mean age of 75.3 + 7.1 years, and the majority were women, white, and had at least some college education. At baseline, there were no significant differences between groups on any demographic or outcome measure. The outcome measures in this study were physical activity, pain, arthritis self-efficacy, and physical function. 


At post-test, pain (as measured by the WOMAC pain scale) decreased in both groups. Peak physical activity measured objectively by wrist-worn accelerometers significantly increased in the AST group at post-test (p = .02). Objective total physical activity and CHAMPS total physical activity tended to increase at post-test in the AST group and decrease in the health education group, but the effects were not statistically significant. 

For the secondary outcomes of arthritis self-efficacy and physical function, no statistically significant effects were found. Arthritis self-efficacy was virtually unchanged at post-test for the AST group and declined slightly in the health education group. Of the physical function measures, 6-minute walk values showed an increase in average walking distance for both groups at post-test.

Main Conclusions

We found that the AST group had higher objective peak activity levels at post-test compared to participants in the health education program. Although peak activity levels from accelerometers may be an indication that people in the AST group were engaging in more intense physical activity, more research needs to be done to examine this. While accelerometers are advantageous because they provide an objective measure of physical activity, it is still not entirely clear what the output means.

With regard to arthritis self-efficacy, we were surprised to find that there was no increase for the AST group. We found that many individuals had positive or negative changes on this variable that may have cancelled out any effects for the group. Further research will be needed to explore the potential mechanisms behind this program.

Implications for Occupational Therapy

The results of this small pilot study were encouraging; however, a larger study is needed to determine if results can be replicated. In addition, it is important to more thoroughly examine underlying mechanisms of treatment. In occupational therapy, we not only need to understand if treatments are effective but also why they work.

The AST provided in this intervention was an adaptation and repackaging of common occupational therapy approaches with a focus on prevention and wellness. Although prevention and wellness is a new area for healthcare service, occupational therapists are particularly suited to help people engage as fully as possible in meaningful and purposeful activities. Patients with knee or hip osteoarthritis are not commonly referred to occupational therapy services because of that condition unless it is to provide rehabilitation after joint replacement surgery.

This study provides preliminary evidence that occupational therapy may be useful to promote symptom management and physical activity engagement among people with osteoarthritis. However, a larger clinical trial will be needed to examine underlying mechanisms of treatment and cost effectiveness. An examination of cost effectiveness is particularly important because it can provide a strong justification for service reimbursement in this area.  
 
It is often difficult to design occupational therapy research because it addresses complex and holistic issues. Materials for the AST program were designed as a detailed session-by-session protocol to reduce interventionist variation and to be reproducible. This type of rigor will be needed to fully test and disseminate interventions.

In order for occupational therapy to thrive as a profession, we need to build our evidence base. Collaboration between practicing clinicians and researchers, as done in this study, is essential to establish services in new practice areas and to optimally address important research questions.

– Susan L. Murphy, ScD, OTR, is assistant professor in the department of Physical Medicine and Rehabilitation at the University of Michigan, as well as a research health science specialist working at the Geriatric Research, Education, and Clinical Center at the Michigan-based VA Ann Arbor Healthcare System.